Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

GLAS

Benchmarks

Task NameDataset NameSOTA ResultTrend
Medical Image SegmentationGlaS (test)
Dice Score93.94
44
Semantic SegmentationGlaS
Dice87
28
Medical Image SegmentationGlaS
Dice92.82
28
Binary SegmentationGLAS
DSC92.35
28
Image ClassificationGlaS
Accuracy98.75
26
Gland SegmentationGlaS (test)
F1 Score91.49
22
Gland SegmentationGlaS Challenge Dataset (test A)
F1 Score92
20
Gland SegmentationGlaS
mIoU0.8684
17
Semantic SegmentationGLaS (test)
mIoU76.06
13
Medical Image SegmentationGlaS (three fixed-seed random data splits)
IoU88.72
8
2D Image SegmentationGlaS
Dice Score83.25
8
Nuclei instance segmentationGlaS (test)
Dice Coefficient72.1
6
SegmentationGlaS (internal held-out)
Dice Score89.3
5
Object DetectionGlaS (testB)
F1 Score73.45
5
Object DetectionGlaS (testA)
F-score0.9039
5
Nuclei ClassificationGlaS transfer from Dpath (test)
Detection Score67.5
5
Lumen SegmentationGlaS Challenge Dataset (test B)
F1 Score71.1
5
Gland SegmentationGlaS (5-fold cross val)
Metric-
0
Showing 18 of 18 rows